GPT-5.6 Luna
Latest Nerf Watch result
On 2026-10-28, GPT-5.6 Luna scored 47.87 on Nerf Watch's suite 2026.10.1, 2.44 points above its 7-day baseline.
- Latest run
- 47.87 on · suite 2026.10.1
- Previous run
- 34.28 on · suite 2026.10.1
- Change from previous run
- +13.59
- 7-day baseline
- 45.43 · 7 baseline runs
- Change from baseline
- +2.44
- Trend
- Higher
A score below the model's own baseline, the mean of its recent runs. One day's dip can be noise, and a lower score is a lower score on this suite, not proof that a provider changed a model.
Score history
As a table
| Day, UTC | Score | Suite |
|---|---|---|
| 47.87 | 2026.10.1 | |
| 34.28 | 2026.10.1 | |
| 18.55 | 2026.10.1 | |
| 46.83 | 2026.10.1 | |
| 50.40 | 2026.10.1 | |
| 54.77 | 2026.10.1 | |
| 47.79 | 2026.10.1 | |
| 65.36 | 2026.10.1 | |
| 44.93 | 2026.10.1 |
Speed
Response time
On , GPT-5.6 Luna's daily response-time probe recorded a median completion time of 2.3 s across three successful responses at high reasoning effort and Standard/default service.
The daily local probe is scheduled at 00:10 UTC. It uses high reasoning effort and three successful fresh single-turn responses to one fixed small task. A day without three successful responses has no point. Lower is faster; dates are UTC.
Drag across the chart to zoom in.Pinch to zoom; drag with two fingers to move.
Output speed
On , GPT-5.6 Luna's Nerf Watch run recorded 120 output tokens per second.
Tokens per second from the daily Nerf Watch run, scheduled at 01:00 UTC: total output tokens divided by total completion time across its four fixed tasks. Effort is low, except DeepSeek V4 Pro uses high, its lowest available variant. Service is Standard/default, or Ultrafast for the two Ultrafast lines. Output includes reported reasoning or thinking tokens once; OpenCode’s separately reported reasoning is added to output. Timing includes waiting within the measured turn and excludes setup and process exit. Higher is faster; dates are UTC.
Drag across the chart to zoom in.Pinch to zoom; drag with two fingers to move.
Facts from the research notes
API price
Latest API price record in Research note 002, as of 2026-10-12: $0.20 input and $1.20 output per million tokens. Record dated 2026-07-30.
developers.openai.com · Research note 002
Context window
No exact-model record in these research notes.
Parameters
No exact-model record in these research notes.